An Eye Movement Research On The Effect Of Contextual Predictability And Word Boundary Information On Chinese Character Position Processing

The reading process involves several cognitive processes such as visual,perceptual,attentional and memory processes,which aim to acquire textual information to achieve comprehension.The interaction model of the reading process suggests that when bottom-up perceptual processing information is insufficient during reading,readers may rely more on top-down information processing to aid reading.When two adjacent Chinese characters in a word whose positions are exchanged or replaced by other Chinese characters may destroy the positional information of the word,resulting in insufficient bottom-up information acquisition,readers may rely on context,a top-down information,to facilitate reading.Therefore,two experiments were conducted to investigate how top-down contextual predictive information and bottom-up word boundary information affect Chinese word position processing in the central and sub-central recesses,respectively.Experiment 1 investigated the effects of contextual predictive and word boundary information on position processing in the central concave,while Experiment 2 explored the effects of contextual predictive and word boundary information on position processing in the subcentral concave.The two experiments were combined to investigate whether contextual predictability had similar effects on location processing at the central and sub-central concave word boundaries.Experiment 1 manipulated the target word presented in the central concave visual field through the boundary paradigm,and examined the effect of contextual predictability on positional processing at word boundaries in central concave processing by varying the type of contextual predictability（high and low）,the type of target word（idiom,four-letter word）,and the type of position（in situ,transposition,and substitution）.It was found that the main effect of location type was significant on first gaze time,gaze time,return gaze time and total gaze time,all of which showed that the in-situ condition was significantly smaller than the replacement condition and the replacement condition was significantly smaller than the substitution condition;there was an interaction between word type and location type on first gaze time,and the first gaze time of the in-situ condition was smaller than the replacement condition when the four-word words were presented,and the in-situ and There was no significant difference between the in situ and transposition conditions for the presentation of idioms.The results suggest that the position coding at word boundaries is stricter in central concave processing,i.e.,permutation and substitution cause longer gaze times.The increase in contextual information did not make the encoding at this location flexible,when bottom-up information processing dominated.Experiment 2,also using the boundary paradigm,manipulated the target words presented in the paracentral concave visual field and examined the effects of contextual predictability and word boundaries on positional processing in paracentral concave processing by varying the type of contextual predictability（high and low）,the type of target words（idioms,four-letter words）,and the type of position（in situ,permutation,and substitution）.The results found that the interaction of the three variables of word type,context type and location type on gaze time and total gaze time was significant,specifically: for when context predictability was low,there was no significant difference in gaze time for four-word words in the transposition and substitution conditions,and there was a significant difference in gaze time for idioms in the transposition and substitution conditions and no significant difference in the transposition and in-situ conditions,i.e.,there was a The permutation effect;when the contextual predictability was high,both four-letter words and idioms showed the permutation effect.The results suggest that top-down contextual predictability plays a role in weakening word boundary information at the late stage of paracentral concave processing.The findings are as follows:(i.In central concave processing,positional processing at word boundaries proceeded in a strict manner,the addition of contextual predictability did not change its encoding,and bottom-up processing of word boundary information dominated.(ii.the enhanced influence of contextual predictability in subcentral concave processing made the positional processing at word boundaries show flexibility in the late stage,and top-down contextual predictability served to weaken the word boundary information.